Development and assessment of a dry eye questionnaire applicable to the Chinese population

Abstract

To develop and assess a new dry eye questionnaire applicable to the Chinese population. Based on literature review and clinical practice, a dry eye questionnaire was developed and optimized to apply to Chinese dry eye patients in the language expression and culture background. Participants (78 patients with dry eye and 82 controls) completed the dry eye questionnaire and the ocular surface disease index (OSDI) questionnaire, and ophthalmic examinations were performed, including slit lamp examination, tear breakup time, fluorescein staining, Schirmer I test and meibomian gland assessment. The original questionnaire was optimized with factor analysis according to the answers from respondents and clinical evaluations. The Cronbach α and intraclass correlation coefficient (ICC) were used to evaluate the internal consistency reliability and test-retest reliability. Factor analysis was used to assess the construct validity, concurrent validity was obtained by Spearman correlation analysis, and discriminant validity was obtained by ANOVA and Wilcoxon rank sum test. Receiver operator characteristics curves were generated to identify the sensitivity and specificity of each questionnaire for diagnosis of dry eye. The questionnaire was optimized to 12 items by factor analysis. The response rate from respondents to the dry eye questionnaire and the OSDI was 100% and 91.25%, respectively. The Cronbachαof the dry eye questionnaire and the OSDI was 0.794 and 0.925, respectively, whilst the ICC of both questionnaires was 0.99, indicating good to excellent reliability. The factor analysis suggested that these two questionnaires had good construct validity. The Spearman correlation analysis indicated that the dry eye questionnaire score correlated positively with the OSDI score (r = 0.812, P < 0.01) and had a greater correlation relationship with the clinical evaluations compared with the OSDI score (r for each was 0.613 and 0.605, P < 0.01). The discriminant validity analysis suggested that there was significant difference in the dry eye questionnaire score between the dry eye group and non-dry eye group (P < 0.01). When the dry eye questionnaire score of 7 was used as the diagnostic threshold, the sensitivity and specificity were 83.33% and 70.73%, respectively, and the area under roc curve was 0.814, which was higher than 0.772 of the OSDI (P < 0.01). The dry eye questionnaire we developed is applicable to the Chinese population with Chinese culture characteristics, high reliability, validity, specificity, and sensitivity, and holds a better diagnostic value than the OSDI for Chinese patients with dry eye.